common

Documentation

Index

• type BatchScheduleDescription
• type FailurePolicy
  • func (f *FailurePolicy) MapToRadixFailurePolicy() *radixv1.RadixJobComponentFailurePolicy
• type FailurePolicyRule
• type FailurePolicyRuleAction
• type FailurePolicyRuleOnExitCodes
• type FailurePolicyRuleOnExitCodesOperator
• type JobScheduleDescription
• type Node
  • func (n *Node) MapToRadixNode() *radixv1.RadixNode
• type RadixJobComponentConfig
• type ResourceList
• type Resources
  • func (r *Resources) MapToRadixResourceRequirements() *radixv1.ResourceRequirements
• type Status
• type StatusReason

Types

type BatchScheduleDescription

type BatchScheduleDescription struct {
	// Defines a user defined ID of the batch.
	//
	// required: false
	// example: 'batch-id-1'
	BatchId string `json:"batchId,omitempty"`

	// JobScheduleDescriptions descriptions of jobs to schedule within the batch
	//
	// required: true
	JobScheduleDescriptions []JobScheduleDescription `json:"jobScheduleDescriptions" yaml:"jobScheduleDescriptions"`

	// DefaultRadixJobComponentConfig default resources configuration
	//
	// required: false
	DefaultRadixJobComponentConfig *RadixJobComponentConfig `json:"defaultRadixJobComponentConfig,omitempty" yaml:"defaultRadixJobComponentConfig,omitempty"`
}

BatchScheduleDescription holds description about batch scheduling job swagger:model BatchScheduleDescription

type FailurePolicy

type FailurePolicy struct {
	// A list of failure policy rules. The rules are evaluated in order.
	// Once a rule matches a job replica failure, the remaining of the rules are ignored.
	// When no rule matches the failure, the default handling applies - the
	// counter of failures is incremented and it is checked against
	// the backoffLimit.
	//
	// maxItems: 20
	// required: true
	Rules []FailurePolicyRule `json:"rules"`
}

FailurePolicy describes how failed job replicas influence the backoffLimit.

func (*FailurePolicy) MapToRadixFailurePolicy

func (f *FailurePolicy) MapToRadixFailurePolicy() *radixv1.RadixJobComponentFailurePolicy

MapToRadixFailurePolicy maps the object to a RadixJobComponentFailurePolicy object

type FailurePolicyRule

type FailurePolicyRule struct {
	// Specifies the action taken on a job replica failure when the onExitCodes requirements are satisfied.
	//
	// required: true
	Action FailurePolicyRuleAction `json:"action"`

	// Represents the requirement on the job replica exit codes.
	//
	// required: true
	OnExitCodes FailurePolicyRuleOnExitCodes `json:"onExitCodes"`
}

FailurePolicyRule describes how a job replica failure is handled when the onExitCodes rules are met.

type FailurePolicyRuleAction

type FailurePolicyRuleAction string

swagger:enum FailurePolicyRuleAction

const (
	// This is an action which might be taken on a job replica failure - mark the
	// job as Failed and terminate all running pods.
	FailurePolicyRuleActionFailJob FailurePolicyRuleAction = "FailJob"

	// This is an action which might be taken on a job replica failure - the counter towards
	// .backoffLimit is not incremented and a replacement replica is created.
	FailurePolicyRuleActionIgnore FailurePolicyRuleAction = "Ignore"

	// This is an action which might be taken on a job replica failure - the replica failure
	// is handled in the default way - the counter towards .backoffLimit is incremented.
	FailurePolicyRuleActionCount FailurePolicyRuleAction = "Count"
)

type FailurePolicyRuleOnExitCodes

type FailurePolicyRuleOnExitCodes struct {
	// Represents the relationship between the job replica's exit code and the
	// specified values. Replicas completed with success (exit code 0) are
	// excluded from the requirement check.
	//
	// required: true
	Operator FailurePolicyRuleOnExitCodesOperator `json:"operator"`

	// Specifies the set of values. The job replica's exit code is checked against this set of
	// values with respect to the operator. The list must not contain duplicates.
	// Value '0' cannot be used for the In operator.
	//
	// minItems: 1
	// maxItems: 255
	// minimum: 0
	// required: true
	Values []int32 `json:"values"`
}

FailurePolicyRuleOnExitCodes describes the requirement for handling a failed job replica based on its exit code.

type FailurePolicyRuleOnExitCodesOperator

type FailurePolicyRuleOnExitCodesOperator string

swagger:enum FailurePolicyRuleOnExitCodesOperator

const (
	// The requirement is satisfied if the job replica's exit code is in the set of specified values.
	FailurePolicyRuleOnExitCodesOpIn FailurePolicyRuleOnExitCodesOperator = "In"

	// The requirement is satisfied if the job replica's exit code is not in the set of specified values.
	FailurePolicyRuleOnExitCodesOpNotIn FailurePolicyRuleOnExitCodesOperator = "NotIn"
)

type JobScheduleDescription

type JobScheduleDescription struct {
	// JobId Optional ID of a job
	//
	// required: false
	// example: 'job1'
	JobId string `json:"jobId,omitempty"`

	// Payload holding json data to be mapped to component
	//
	// required: false
	// example: {'data':'value'}
	Payload string `json:"payload"`

	// RadixJobComponentConfig holding data relating to resource configuration
	//
	// required: false
	RadixJobComponentConfig `json:",inline"`
}

JobScheduleDescription holds description about scheduling job swagger:model JobScheduleDescription

type Node

type Node struct {
	// Defines rules for allowed GPU types.
	//
	// required: false
	Gpu string `json:"gpu,omitempty"`

	// Defines minimum number of required GPUs.
	//
	// required: false
	GpuCount string `json:"gpuCount,omitempty"`
}

Node defines node attributes, where container should be scheduled

func (*Node) MapToRadixNode

func (n *Node) MapToRadixNode() *radixv1.RadixNode

MapToRadixNode maps the object to a RadixNode object

type RadixJobComponentConfig

type RadixJobComponentConfig struct {
	// Resource describes the compute resource requirements.
	//
	// required: false
	Resources *Resources `json:"resources,omitempty"`
	// Node defines node attributes, where container should be scheduled
	//
	// required: false
	Node *Node `json:"node,omitempty"`

	// TimeLimitSeconds defines maximum job run time. Corresponds to ActiveDeadlineSeconds in K8s.
	//
	// required: false
	TimeLimitSeconds *int64 `json:"timeLimitSeconds,omitempty"`

	// BackoffLimit defines attempts to restart job if it fails. Corresponds to BackoffLimit in K8s.
	//
	// required: false
	BackoffLimit *int32 `json:"backoffLimit,omitempty"`

	// FailurePolicy defines how failed job replicas influence the backoffLimit.
	//
	// required: false
	FailurePolicy *FailurePolicy `json:"failurePolicy,omitempty"`

	// ImageTagName defines the image tag name to use for the job image
	//
	// required: false
	ImageTagName string `json:"imageTagName,omitempty"`
}

RadixJobComponentConfig holds description of RadixJobComponent

type ResourceList

type ResourceList map[string]string

type Resources

type Resources struct {
	// Limits describes the maximum amount of compute resources allowed.
	//
	// required: false
	Limits ResourceList `json:"limits,omitempty"`

	// Requests describes the minimum amount of compute resources required.
	// If Requests is omitted for a container, it defaults to Limits if
	// that is explicitly specified, otherwise to an implementation-defined value.
	//
	// required: false
	Requests ResourceList `json:"requests,omitempty"`
}

Resources describes the compute resource requirements.

func (*Resources) MapToRadixResourceRequirements

func (r *Resources) MapToRadixResourceRequirements() *radixv1.ResourceRequirements

MapToRadixResourceRequirements maps the object to a ResourceRequirements object

type Status

type Status struct {

	// Status of the operation.
	// One of: "Success" or "Failure".
	// example: Failure
	Status string `json:"status,omitempty"`
	// A human-readable description of the status of this operation.
	// required: false
	// example: job job123 is not found
	Message string `json:"message,omitempty"`
	// A machine-readable description of why this operation is in the
	// "Failure" status. If this value is empty there
	// is no information available. A Reason clarifies an HTTP status
	// code but does not override it.
	// required: false
	// example: NotFound
	Reason StatusReason `json:"reason,omitempty"`

	// Suggested HTTP return code for this status, 0 if not set.
	// required: false
	// example: 404
	Code int `json:"code,omitempty"`
}

Status is a return value for calls that don't return other objects or when a request returns an error swagger:model Status

type StatusReason

type StatusReason string

const (
	StatusSuccess = "Success"
	StatusFailure = "Failure"
	// StatusReasonUnknown means the server has declined to indicate a specific reason.
	// Status code 500.
	StatusReasonUnknown StatusReason = "InternalError"
	// StatusReasonBadRequest means that the operation could not be fulfilled.
	// Status code 400
	StatusReasonBadRequest StatusReason = "BadRequest"
	// StatusReasonNotFound means one or more resources required for this operation
	// could not be found.
	// Status code 404
	StatusReasonNotFound StatusReason = "NotFound"
	// StatusReasonInvalid means the requested create or update operation cannot be
	// completed due to invalid data provided as part of the request. The client may
	// need to alter the request.
	// Status code 422
	StatusReasonInvalid StatusReason = "Invalid"
)